Second, routing
       Alignment of current density: now the majority of electronic circuit board using insulated copper constitute tied. Common PCB copper thickness of 35μm, the alignment value can be obtained in accordance with 1A/mm experience the value of current density, the specific calculations can be found in textbooks. To ensure the alignment principles of mechanical strength should be greater than or equal to the width of 0.3mm (other non-power supply circuit board may be smaller minimum line width). PCB copper thickness of 70μm is also common in switching power supply, then the current density can be higher.
       Add that, now Changyong circuit board design tool design software generally items such as line width, line spacing, hole size and so dry plate Guo Jin Xing parameters can be set. In the design of circuit boards, design software automatically in accordance with the specifications, can save time, reduce some of the workload and reduce the error rate.
       Generally higher on the reliability of lines or line density wiring can be used double panel. Characterized by moderate cost, high reliability, to meet most applications.
       The ranks of some of the power module products are also used plywood, mainly to facilitate integration of power devices such as transformer inductance to optimize wiring, cooling and other power tube. Good consistency with the craft beautiful, transformer cooling good advantage, but its disadvantage is high cost, poor flexibility, only suitable for industrial mass production.
       Single-sided, the market circulation of almost universal switching power supply using single-sided circuit board, which has the advantage of lower costs in the design and production technology are also taken some measures to ensure its performance.
Single PCB design today to talk about some experience, as a single panel with low cost, easy-to-manufacture features, the switching power supply circuit has been widely used, because of its side tied only copper, the device's electrical connections, mechanical fixation should rely on the copper layer, the processing must be careful.
       To ensure good performance of the mechanical structure welding, single-sided pad should be slightly larger to ensure that the copper and substrate tied good focus, and thus will not be shocked when the copper strip, broken off. General welding ring width should be greater than 0.3mm. Pad diameter should be slightly larger than the diameter of the device pins, but not too large, to ensure pin and pad by the solder connection between the shortest distance, plate hole size should not hinder the normal conditions for the degree of investigation, the pad diameter is generally greater than pin diameter 0.1-0.2mm. Multi-pin device to ensure a smooth investigation documents can also be larger.
       Electrical connection should be as wide as possible, in principle, should be larger than the width of pad diameter, special circumstances should be connected in line with the need to widen the intersection pad (commonly known as Generation tears), to avoid breaking certain conditions, line and pad. Principle of minimum line width should be greater than 0.5mm.
       Single-board components to be close to the circuit board. Need overhead cooling device to device and circuit board between the pins plus casing, can play a supporting device and increase the dual role of insulation to minimize or avoid external shocks on the pad and the pin junction impact and enhance the firmness of welding. Circuit board supporting the weight of large parts can increase the connection point, can enhance joint strength between the circuit board, such as transformers, power device heat sink.
       Single-sided welding pins without affecting the surface and the shell spacing of the prior conditions, it can be to stay longer, the advantage of increased strength of welded parts, increase weld area and immediately found a Weld phenomenon. Shear pin long legs, the welding force smaller parts. In Taiwan, the Japanese often use the device pins in the welding area and the circuit board was bent 45 degrees, and then welding process, its reasoning Ibid. Double panel today to talk about the design of some of the issues, in relatively high number of requests, or take the line density of the larger application environments using double-sided PCB, its performance and various indicators of a lot better than a single panel.
       Two-panel pad as holes have been high intensity metal processing, welding ring smaller than a single panel, the pad hole diameter slightly larger in diameter than pins, as in the welding process solder solution conducive to penetrate through the top hole solder pad to increase the welding reliability. But there is a disadvantage if the hole is too large, wave soldering tin when the jet impact in the lower part of the device may go up, have some flaws.
       High current handling of alignment, line width in accordance with pre-quote processing, such as the width is not enough to go online in general can be used to increase the thickness of tin plating solution, the method has a good variety of
       1. Will take the line set to pad property, so that when the circuit board manufacturing solder alignment will not be covered, the whole hot air normally be tin plated.
       2. In the wiring by placing pads, the pad is set to take in line shape, pay attention to the pad holes set to zero.
       3. In the solder layer placed on line, this method is the most flexible, but not all PCB manufacturers will understand your intentions, needed captions. Place the line in the solder layer of the site will not coated solder tinning line several methods as above, to note that, if the alignment of a very wide all plated with tin in solder after the solder will bond a lot and distribution is very uneven, affecting appearance. Article tin can be used generally slender width in the 1 ~ 1.5mm, length can be determined according to lines, tin part of the interval 0.5 ~ 1mm
       Double-sided circuit board for the layout, the alignment provides a very selective, make wiring more reasonable. On the ground, the power ground and signal ground must be separated, the two to converge in filter capacitors, in order to avoid a large pulsed current through the signal ground connection instability caused by unexpected factors, the signal control circuit grounding point as far as possible, a skill, as far as possible the alignment of the non-grounded wiring layer in the same place, the last shop in another layer of earth. Output line through the filter capacitors, the general first, and then to the load, input line must also pass capacitor, to the transformer, the theoretical basis is to ripple through trip filter capacitor.
       Voltage feedback sampling, in order to avoid high current through the alignment of the feedback voltage on the sampling point must be the most peripheral power output to increase the load effect of target machine.
       Alignment change from a wiring layer to another wiring layer generally used hole connected, not through the pin pad device to achieve, because the plug in the device may be damaged when the relationship between this connection, there is current in every passage of 1A, at least two through-hole, through hole diameter is greater than the principle of 0.5mm, 0.8mm generally processed ensure reliability.
       Cooling devices, in some small power supply, the circuit board traces can be and cooling, characterized by the alignment as generous as possible to increase the cooling area is not coated solder, conditions can even be placed over holes, enhanced thermal conductivity .
       Today to talk about the aluminum plate in the switching power supply application and multilayer printed circuit in the switching power supply applications.
       Aluminum plate by its own structure, has the following characteristics: very good thermal conductivity, single Mianfu copper, the device can only be placed in tied copper surface, can not open electrical connection hole so as not to place jumper in accordance with a single panel.
       Aluminum plate is generally placed patch device, switch, the output rectifier heat conduction through the substrate to go out, very low thermal resistance, high reliability can be achieved. Transformer with planar chip structure, but also through substrate cooling, the temperature is lower than the conventional, the same size transformer with a large aluminum plate structure available output power. Aluminum plate jumper bridge approach can be used. Aluminum plate power are generally composed by the two PCB, another one to place the control circuit board, through the physical connection between the two boards is integrated.
       As the excellent thermal conductivity of aluminum plate, in a small amount of manual welding more difficult, solder cooling too fast and prone to problems of a simple and practical way of existing, an ironing ordinary iron (preferably temperature regulation function), over and iron for the last, fixed, and temperature to 150 ℃ and above the aluminum plate on the iron, heating time, and then affix the components according to conventional methods and welding, soldering iron temperature is appropriate to the device easy to , is too high when the device may be damaged, or even copper strip aluminum plate, the temperature is too low welding effect is not good, to be flexible.
       Recent years, with the multi-layer circuit board applications in switching power supply circuit, printed circuit transformer makes it possible, due to multilayer, smaller spacing also can take advantage of Bianya Qi window section, the main circuit board can be re- Add 1-2 formed by the multilayer printed coil to use the window, the purpose of reducing circuit current density, due to adopt printed coil, reducing manual intervention, transformers consistency, surface structure, low leakage inductance, coupling good . Open-type magnetic core, good heat dissipation. Because of its many advantages, is conducive to mass production, it is widely used. But the research and development of large initial investment, not suitable for small-scale health.
       Switching power supply is divided into, two forms of isolation and non-isolated, isolated here mainly to talk about switching power supply topologies form below, non-specified, are to isolate the power. Isolated power supply in accordance with the structure of different forms, can be divided into two categories: a forward and flyback. Flyback transformer primary side means that when the Vice-edge conduction cut-off, transformer storage. Close of the primary, secondary side conduction, the energy released to the load of work status, general conventional flyback power multiplex, twin-tube is not common. Forward refers to the primary conduction in transformer secondary side while the corresponding output voltage is induced into the load, the direct transfer of energy through the transformer. According to specifications can be divided into conventional forward, including the single-transistor forward, Double Forward. Half-bridge, bridge circuits are all forward circuit.
       Forward and flyback circuits have their own characteristics in the process of circuit design to achieve optimal cost-effective, can be applied flexibly. Usually in the low-power flyback can be adopted. Slightly larger forward circuit can use a single tube, medium-power can use Double Forward circuit or half-bridge circuit, low-voltage push-pull circuit, and the half-bridge work in the same state. High power output, generally used bridge circuit, low voltage can be applied push-pull circuit.
Flyback power supply because of its simple structure, and to cut the size of a similar size and transformer inductance, the power supply in the medium has been widely applied. Presentation referred to in some flyback power supply can do dozens of watts, output power exceeding 100 watts would be no advantage to them difficult. Under normal circumstances, I think so, but it can not be generalized, PI's TOP chips can do 300 watts, an article describes the flyback power supply can be on the KW, but not seen in kind. Power output and the output voltage level.
       Flyback power transformer leakage inductance is a critical parameter, because the power needs of the flyback transformer stored energy, to make full use of transformer core, the general must be open in the magnetic circuit air gap, the aim is to change the core hysteresis back line of the slope, so that transformers can withstand the impact of a large pulse current, which is not core into saturation non-linear state, the magnetic circuit in the high reluctance air gap in the state, generated in the magnetic flux leakage is much larger than completely closed magnetic circuit .
       Transformer coupling between the first pole is the key factor determining the leakage inductance, the coil to be very close as far as possible the first time, the sandwich can be used around the law, but this would increase the distributed capacitance transformer. Use core as core with a long window, can reduce the leakage inductance, such as the use of EE, EF, EER, PQ-based EI type magnetic core effective than good.
       The duty cycle of flyback power supplies, in principle, the maximum duty cycle of flyback power supply should be less than 0.5, otherwise not easy loop compensation may be unstable, but there are some exceptions, such as the U.S. PI has introduced the TOP series chip can work under the conditions of duty cycle is greater than 0.5.
       Duty cycle by the transformer turns ratio to determine former deputy side, I am an anti-shock view is, first determine the reflected voltage (output voltage reflected through the transformer coupling the primary voltage value), reflecting a certain voltage range of voltage increase is duty cycle increases, lower power loss. Reduce the reflected voltage duty cycle decreases, increases power loss. Of course, this is a prerequisite, when the duty cycle increases, it means that the output diode conduction time, in order to maintain output stability, more time will be to ensure that the output capacitor discharge current, the output capacitor will be under even greater high-frequency ripple current erosion, while increasing its heat, which in many circumstances is not allowed.
       Duty cycle increases, change the transformer turns ratio, transformer leakage inductance will increase, its overall performance change, when the leakage inductance energy large enough, can switch to fully offset the large account space to bring low-loss, no further increase when the meaning of duty, because the leakage inductance may even be too high against the peak voltage breakdown switch. Leakage inductance as large, may make the output ripple, and other electromagnetic indicators deteriorated. When the duty hours, the high RMS current through the switch, transformer primary current rms and lowered the converter efficiency, but can improve the working conditions of the output capacitor to reduce fever. How to determine the transformer reflected voltage (duty cycle)
       Some netizens said switching power supply feedback loop parameter settings, work status analysis. Since high school mathematics is rather poor, "Automatic Control Theory," almost on the make-up, and for the door is still feeling fear, and now can not write a complete closed-loop system transfer function, zero for the system, the concept of feeling pole vague, see Bode plot is only about to see is a divergence or convergence, so the feedback compensation can not nonsense, but there are a number of recommendations. If you have some mathematical skills, and then have some time to learn then the University of textbooks, "Principles of Automatic Control" digest look carefully to find out, combined with practical switching power supply circuit, according to the work of state for analysis. Will be harvested, the Forum has a message, "coach feedback loop to study the design, debugging," in which CMG good answer, I think we can reference.
       Then today, on the duty cycle of flyback power supply (I am concerned about the reflected voltage, consistent with the duty cycle), the duty cycle with the voltage selection switch is related to some early flyback switching power supply using a low pressure tube, such as 600V or 650V AC 220V input power as a switch, perhaps when the production process, high pressure tubes, easy to manufacture, or low-pressure pipes are more reasonable conduction losses and switching characteristics, as this line reflected voltage can not be too high, otherwise the work order to switch the security context of loss of power absorbing circuit is quite impressive.
       Reflected voltage 600V tube proved not more than 100V, 650V tube reflected voltage not greater than 120V, the leakage inductance voltage spike when the tubes are clamped at 50V 50V working margin. Now that the MOS raise the level of manufacturing process control, flyback power supplies are generally used 700V or 750V or 800-900V the switch. Like this circuit, overvoltage capability against a number of switching transformer reflected voltage can be done a bit higher, the maximum reflected voltage in the 150V is appropriate, to obtain better overall performance.
       TOP PI's recommendation for the 135V chipset with transient voltage suppression diode clamp. But his evaluation board generally reflected voltage to be lower than the value at around 110V. Both types have their advantages and disadvantages:
Category: shortcomings against over-voltage, low duty cycle is small, a large pulse current transformer primary. Advantages: small transformer leakage inductance, electromagnetic radiation and low ripple index higher switch loss, the conversion efficiency is not necessarily lower than the second.
       The second category: a large number of shortcomings of power loss, a large number of transformer leakage inductance, the ripple worse. Advantages: Some strong against over-voltage, large duty cycle, lower transformer losses and efficiency higher.
       Reflected voltage flyback power supply and a determining factor
       Reflected voltage flyback power supply with a parameter related to that is the output voltage, output voltage, the lower the larger the transformer turns ratio, the greater the transformer leakage inductance, switch to withstand higher voltage breakdown switch is possible to absorb power consumption is higher, has the potential to permanently absorb the circuit power device failure (particularly with transient voltage suppression diode circuits). In the design of low-voltage low-power flyback power output optimization process must be handled with care, its approach has several:
       1, using a large core of a power level lower leakage inductance, which can improve the low-voltage flyback power conversion efficiency, reduce losses, reduce output ripple and improve multi-output power of the cross regulation in general is common in household appliances with a switch power, such as CD-ROM drive, DVB set-top boxes.
       2, if the conditions were not increased core, can reduce the reflected voltage, reducing the duty cycle. Reduce the reflected voltage can reduce the leakage inductance but may reduce the power conversion efficiency, which is a contradiction between the two, must have an alternative process to find a suitable point, replace the transformer during the experiment can detect the transformer original side of the anti-peak voltage, peak voltage to minimize the anti-pulse width, and magnitude of the work safety margin increase converter. Generally reflected voltage 110V when appropriate.
       3, enhance the coupling, reducing losses, the introduction of new technologies, and the routing process, transformers to meet the security specifications will between the primary and secondary side to insulation measures, such as pad tape, plus side air insulation tape. These will affect the performance of transformer leakage inductance, the reality can be used in production around the primary winding secondary wrapping method. Or sub-system with a triple insulated wire wound to remove the insulation between the initial level, can enhance the coupling, even use wide copper winding.
       The article refers to low voltage output is less than or equal to 5V output, as this type of small power supply, my experience is that the power output of more than 20W output can use a forward, get the best value for money, of course, this is not the right decision , and personal habits, relationship between the application environment, the next time to talk about the flyback power supply with a magnetic core, magnetic circuit air gap opening some understanding, I hope you receive adequate guidance.
              Flyback power transformer core magnetization state at work in one way, it needs to open the air gap magnetic circuit, similar to the pulsating direct current sensor. Part of the magnetic coupling through the air gap. Why I understand the principle of open air gap as follows: As the power ferrite also has a similar rectangle of the operating characteristics (hysteresis loop), operating characteristics curve in the Y-axis magnetic induction (B), now the general production process saturation point in 400mT above, the general value in the design of this value should be more appropriate in the 200-300mT, X-axis magnetic field strength (H) the value of current intensity is proportional to the magnetization. Open magnetic circuit air gap equal to the magnetic hysteresis loop to the X axis tilt, in the same magnetic induction intensity, can withstand a greater magnetizing current, equivalent to core store more energy, this energy cut-off switch When spilled into the load through the transformer secondary circuit, flyback power core to open the air gap is twofold. One is to transfer more energy, and the second to prevent the core into saturation.
       Flyback Power Transformer magnetization state in one way, not only to pass through the magnetic coupling energy, is also responsible for input and output isolation voltage transform multiple roles. Therefore, the treatment gap need to be very careful, the air gap leakage inductance can become too large, increase the hysteresis loss, iron loss, copper loss increases, affecting the power of the whole performance. Air gap is too small has the potential to transformer core saturation, resulting in damage to power
       The so-called flyback power supply is continuous and discontinuous mode transformer working conditions, working in full load condition in the power transformer complete transfer, or incomplete transmission mode. General design of the working environment, conventional flyback power supply should work in continuous mode, this switch, circuit loss are relatively small, and can reduce the stress of work input and output capacitors, but that there are some exceptions.
       Requires in particular that: As the characteristics of the flyback power supply is also more suitable for design into a high-voltage power supply, and high-voltage power transformers generally work in discontinuous mode, I understand the need for as high voltage power supply output voltage of the rectifier diodes. Because of the manufacturing process characteristics, high-tension diode, reverse recovery time is long, low speed, the current continuous state, the diode has a positive bias in the recovery, reverse recovery energy loss is very large, is not conducive to converter performance increase, ranging from reduced conversion efficiency, rectifiers, severe fever, weight is even burnt rectifier. As in the intermittent mode, the diode is reverse biased under zero bias, loss can be reduced to a relatively low level. Therefore, high voltage power supply work in discontinuous mode, and the frequency can not be too high.
       Another type of flyback power supply work in the critical state, the general type of power supply work in FM, or FM-width-modulated dual-mode, a number of low-cost self-excitation power (RCC) is often used this form in order to ensure stable output transformer As the operating frequency, output current or input voltage change, close to the fully loaded transformer is always maintained at between continuous and intermittent, this power is only suitable for small power output, otherwise the handling characteristics of electromagnetic compatibility will be a headache
       Flyback switching power supply transformer should work in continuous mode, it required relatively large winding inductance, of course, is to some extent continuous, excessive pursuit of absolute continuity is not realistic, may need a great core, very much coil turns, accompanied by a large leakage inductance and distributed capacitance, worth the trouble. So how does this parameter to determine, through repeated practice, and analysis of peer design, I think, in the nominal voltage input, the output reached 50% and 60% transformer from intermittent, continuous state of transition to more appropriate. Or at the highest input voltage state, the full output, the transformer can transition to the continuous state on it.